Artificial intelligence-driven three-dimensional modeling of biliary strictures using magnetic resonance cholangiopancreatography for quantitative assessment and stent planning

doi:10.3748/wjg.125661

Advanced Search

BPG is committed to discovery and dissemination of knowledge

Home / Archive / Volume 33, Issue 2

This Article

(0)

(7)

Peer-Review Report of This Article

Supplementary Materials of This Article

Journal Information of This Article

Publication Name

World Journal of Gastroenterology

ISSN

1007-9327

Publisher of This Article

Baishideng Publishing Group Inc, 7041 Koll Center Parkway, Suite 160, Pleasanton, CA 94566, USA

- Full Article (XML)

Observational Study

Copyright: ©Author(s) 2026. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution-NonCommercial (CC BY-NC 4.0) license. No commercial re-use. See permissions. Published by Baishideng Publishing Group Inc.

World J Gastroenterol. Jan 14, 2027; 33(2): 125661
Published online Jan 14, 2027. doi: 10.3748/wjg.125661

Artificial intelligence-driven three-dimensional modeling of biliary strictures using magnetic resonance cholangiopancreatography for quantitative assessment and stent planning

Ji-Hun Chun, Jin-Woo Hong, Jong-Uk Hou, Kyong Joo Lee, Da Hae Park, Hye Won Cha, Se Woo Park, Seon Jeong Min, Wooyoung Yu

Ji-Hun Chun, Jin-Woo Hong, Jong-Uk Hou, Division of Software, Hallym University, Chuncheon 24252, South Korea

Kyong Joo Lee, Da Hae Park, Hye Won Cha, Se Woo Park, Division of Gastroenterology, Department of Internal Medicine, Hallym University Dongtan Sacred Heart Hospital, Hallym University College of Medicine, Hwaseong 18450, Gyeonggi, South Korea

Seon Jeong Min, Wooyoung Yu, Department of Radiology, Hallym University Dongtan Sacred Heart Hospital, Hallym University College of Medicine, Hwaseong 18450, Gyeonggi-do, South Korea

Co-first authors: Ji-Hun Chun and Jin-Woo Hong.

Co-corresponding authors: Jong-Uk Hou and Se Woo Park.

Author contributions: Chun JH and Hong JW designed and performed the research, developed the software, analyzed the data, and wrote the manuscript, contributed equally to this work as co-first authors; Hou JU and Park SW conceptualized and supervised the study, contributed to the methodology and project administration, and revised the manuscript, contributed equally to this work as co-corresponding authors; Lee KJ, Park DH, and Cha HW provided clinical resources, curated data, and performed investigation and validation; Park SW conceptualized and supervised the study, provided resources and project administration, and revised the manuscript; Min SJ and Yu W provided radiological resources, data curation, and validation; all authors have read and approved the final version of the manuscript and agree to be accountable for all aspects of the work.

AI contribution statement: During the preparation and revision of this manuscript, the authors used ChatGPT (OpenAI, San Francisco, CA, United States) solely as a limited assistive tool for English-language refinement and cross-checking linguistic clarity, grammatical consistency, and internal consistency of the manuscript. The AI tool was not used to generate research data, perform statistical analyses, create or manipulate images, interpret the study findings, formulate scientific conclusions, or make clinical or scientific decisions. All AI-assisted suggestions were critically reviewed, independently verified, and revised as appropriate by the authors. The authors take full responsibility for the originality, accuracy, scientific integrity, and final content of the manuscript.

Supported by the National Research Foundation of Korea (NRF) grant funded by the Korean Government (Ministry of Science and ICT, MSIT), No. RS-2026-25477876; and the Hallym University Medical Center Research Fund (Mighty Hallym 4.0).

Institutional review board statement: This study was reviewed and approved by the Institutional Review Board of Hallym University Dongtan Sacred Heart Hospital (approval No. 2023-09-003-001), and was conducted in accordance with the ethical guidelines of the Declaration of Helsinki.

Informed consent statement: This was a retrospective study using de-identified, clinically acquired imaging data; the requirement for written informed consent was waived by the Institutional Review Board of Hallym University Dongtan Sacred Heart Hospital. All data were anonymized to protect patient privacy.

Conflict-of-interest statement: All authors declare no conflict of interest in publishing the manuscript.

STROBE statement: The authors have read the STROBE Statement – checklist of items, and the manuscript was prepared and revised according to the STROBE Statement – checklist of items.

Data sharing statement: All data relevant to this study are included in the article and its Supplementary material. Additional extracted data are available from the corresponding author on reasonable request.

Corresponding author: Se Woo Park, MD, PhD, Professor, Division of Gastroenterology, Department of Internal Medicine, Hallym University Dongtan Sacred Heart Hospital, Hallym University College of Medicine, 7 Keunjaebong-gil, Hwaseong 18450, Gyeonggi, South Korea. britnepak@hallym.or.kr

Received: July 14, 2026
Revised: August 5, 2026
Accepted: September 17, 2026
Published online: January 14, 2027
Processing time: 116 Days and 17.1 Hours

Abstract

BACKGROUND

Magnetic resonance cholangiopancreatography (MRCP) depicts the biliary tree three-dimensionally but is used mainly for qualitative diagnosis rather than quantitative procedural planning. Endoscopic drainage of biliary strictures, particularly hilar and intrahepatic lesions, depends on an accurate understanding of stricture location, length, ductal curvature, and branch involvement, yet MRCP information is seldom converted into reproducible measurements. Artificial intelligence (AI) may bridge this gap by transforming diagnostic imaging into quantitative, procedure-oriented geometry. We hypothesized that MRCP-derived biliary anatomy could be converted into an interpretable three-dimensional (3D) model supporting biliary stricture assessment and candidate stent-planning geometry.

AIM

To develop and evaluate a proof-of-concept AI workflow that converts MRCP-derived biliary anatomy into a quantitative 3D stricture model.

METHODS

This single-center study analyzed 384 MRCP examinations (242 controls and 142 with suspected biliary stricture). Expert-corrected masks and 3D annotations were used to train or evaluate modules for region-of-interest localization, bile duct segmentation, projection-based multi-view classification, synthetic morphology restoration, and centerline-guided tubular mesh generation. Binary stricture detection and three-class severity classification were assessed with patient-level cross-validation after excluding unavailable or task-incompatible labels. Segmentation and restoration were evaluated using Dice scores.

RESULTS

In 380 evaluable cases, projection-based binary classification achieved an accuracy of 0.9770, balanced accuracy of 0.9721, macro F1 score of 0.9749, and receiver operating characteristic area under the curve of 0.9953. Three-class severity classification was less accurate (accuracy 0.8711), indicating that severity stratification is more difficult than binary detection. The best Dice score was 0.8113 in the segmentation comparison and 0.8944 in the extended normal-group experiment. In synthetic restoration, DiceFocalLoss produced the highest Dice score (0.7531), and lesion weighting improved reconstruction of interrupted duct segments. In anatomically simple cases, restored geometry was converted into a centerline-guided tubular mesh, whereas complex hilar or intrahepatic branching remained vulnerable to projection overlap.

CONCLUSION

MRCP-based AI modeling of biliary strictures is feasible for quantitative 3D assessment and exploratory stent-planning geometry. External validation and prospective clinical evaluation are required before procedural implementation.

Keywords: Biliary stricture; Magnetic resonance cholangiopancreatography; Artificial intelligence; Three-dimensional reconstruction; Stent planning

Core Tip: This proof-of-concept study shows that magnetic resonance cholangiopancreatography-derived biliary anatomy can be transformed by an integrated artificial-intelligence workflow into a quantitative three-dimensional model of biliary strictures. By combining region-of-interest localization, transformer-based segmentation, projection-based multi-view classification, synthetic morphology restoration, and centerline-guided mesh generation, the pipeline reframes biliary stricture analysis from qualitative image review toward reproducible structural modeling and candidate stent-planning geometry. Binary stricture detection was highly accurate internally, whereas reconstruction of complex hilar or intrahepatic anatomy remained challenging and requires external validation before clinical use.